Octaver #2 help needed

[Mr. Lime]

Some time ago I wanted to build an octaver/overdrive pedal based on CMOS similar to the Red Lama Fuzz.
Things got too complicated and so I would like to give it another try with op-amps instead.

Once again I'm looking for the DBA Octave Clang sound with some minor tweaks.
I took an active James tone stack from the Xotic AC Booster at the end of the circuit. In front of the tone stack is a recovery/boost stage.

The overdrive part should be switchable from TS like soft clipping to OCD style hard clipping.

I'm not sure if the filtering part is alright, I think a coupling cap somewhere between R3 and the blend pot is missing.

I would be glad for any suggestion..
Thanks a lot!

[PRR]

A lot going on there.

What does IC1b do?

Why does C2 go to Vref?

Why does Diode SW go to ground?

How come Octave SW1 doesn't break the path from the Blend pot wiper? (Ah, Octave SW2 may indeed kill the octave path.)

Why does the transformer primary return to IC1a "- In"? It appears to defeat the point of Fuzz pot. Also it will thump when Octave SW2 is thrown.

Why are IC2a and IC2b "+ Ins" tied to ground?

What is R11 for??

One of R7 or R8 is not needed and not wanted; I would have to think or research to remember which; I leave that for you.

[Mr. Lime]

Hi and thanks for replying!

I must have been pretty tired when I draw this.
Sure, the IC + Inputs should go to Vref instead of ground.

IC1b is a buffer, I'm not sure if I can ditch it?

The diode SW switchs between soft- and hard clipping.

R7 and R8 are something like a fixed "shift" control, I may replace it with a 50k trim pot, I have seen this in Merlin Blencowe valve preamps book.

R11 is removed..
I forgot a coupling cap from the Fuzz pot to Octave SW2, than it should work?

The rest is based on OCD, Maxon OD808 and the attached schematic.
Thanks a lot!

[Mr. Lime]

Here's the updated circuit.

The input buffer is now based on Harald Sabro's "2in1_od_boost" Klon input buffer,
I removed R11 and replaced R7/R8 with a trim pot.

Missed to rename the James TS controls for treble and bass.

Anything else that looks weird?
I'm thankful for any suggestion.  

Thanks

[Mr. Lime]

Here's my second version:

I replaced the James tone stack with a BMP TS and added a mid cut/boost circuit in the input stage to pre-emphasize the mid range that isn't controlled by the BMP TS.
I'm not sure if lifting the mid circuit's ground will remove it's influence on the bypass mode.

Anything else that should be changed?

Thanks in advance.

[Mr. Lime]

Just to make sure before I start building this, the octaver part with the transformer looks good now?

Thanks

[Cozybuilder]

Have you tried breadboarding this yet?

[Mr. Lime]

No, not yet cause some parts are missing..

I already ordered them, meanwhile I'm looking for help to find potential mistakes as I'm not sure about these untypical circuit parts like the transformer or the active mid control at the input stage.

To give an input:

In bypass mode, R23 is taken from ground which means C20 and C21 bypass the mid pot with 6nF. R.G writes on geofex that the 51pF across the diodes in the tube screamer acts to soften the distortion, most noticeable when the 500k gain pot is maxed out.

So how would the 6nF effect the 100k pot in the first stage?

Pretty much the same question for the second stage where C5 and C14 would cross the 1M Fuzz pot with 9nF
They are also parallel with C1 100pF.
What's is the effect of it?

Thanks

[Mr. Lime]

To answer my question:

C20 and C21 form a low pass with the 100k pot, which would kill all the highs and mids at 6n // 100k.
So lifting R23 won't work.

Now how can I use the input stage as a buffer in bypass mode?

Maybe it's better to leave the mid control away and just use the buffer.
..or keeping the mid control active in bypass..

[Mr. Lime]

Here's my new idea:

I changed the topology to buffer (Klon) -> tone control -> gain stage -> octave blend -> recovery stage with mid control

The gain stage looks different too now, I would like to try out the "Tube Type Distortion" circuit from AMZ.
I tryed to keep it more simple and ditched the hard clipping stuff, the octave switch and the fat switch.

Power supply is changed to a IC7660 to archive higher output swing for buffer and gain stage.

Is there anywhere a mistake that I have left over?
Coupling caps and resistors to ground look good?

Thanks a lot.
If everything is fine, I would start create a Layout that other people can try this circuit too.

[Cozybuilder]

Just a quick glance-
Why are Di and D2 back to back?
I think your blend pot will have a massive dead spot in the middle- you might want to try something on the order of 10K - 50K instead.
Why are the + inputs tied to ground, not Vref (usually V+/2)?
Why not use M for 1000K up?

I'm sure we look forward to hearing how it works once you've had a chance to breadboard it- that's half the fun of this hobby.

[Mr. Lime]

Why are Di and D2 back to back?

I just draw the circuit part found on: http://www.muzique.com/schem/projects.htm
Don't know exactly how it works but I'm sure it's interesting.

I think your blend pot will have a massive dead spot in the middle- you might want to try something on the order of 10K - 50K instead.

Good point! In the japanese schematic some posts before, a 1M pot is used but 25k or 50k might be the better choice, you're right.

Why are the + inputs tied to ground, not Vref (usually V+/2)?

The power supply with the charge pump creates +9V/-9V (=18V) so ground is already "virtual ground" because of the split supply.
This causes less parts and a simple layout.

Why not use M for 1000K up?

You mean for the mid pot?
I got the active mid control idea from here: http://loudifier.com/projects/guitar-bass-overdrive/
He uses 100k too.

[highwater]

Why are Di and D2 back to back?

I just draw the circuit part found on: http://www.muzique.com/schem/projects.htm
Don't know exactly how it works but I'm sure it's interesting.

It's because they're zener diodes - they conduct at the zener voltage in one direction, and as a normal diode in the other. As drawn, they will clip at ~4v (3.3v zener plus ~0.7 normal diode) in one direction and ~7.5v in the other direction.

[Mr. Lime]

Why not use M for 1000K up?

Ah, you mean "Mega". My drawing software does not take M instead..

Here's my update:

I added a switch (S2) to switch between the overdrive signal and the clean signal blend.

Does mixing the clean channel work this way?
Anything else that looks strange?

[Cozybuilder]

Assuming you are using + and - power supply to the op-amps, I think its ready to breadboard.

[Mr. Lime]

Alright, what transformer should I take?

I would give the TM013 a shot.

Transformer with a primary impedance of 1k and a secondary impedance of 8R, turn ratio is 11.486:1.

The primary has an inductance of about 500mH, it can be used as a very low cost wah inductor!

Or should i order something similar to the Tycobrahe Octavia 42TM022?

[Cozybuilder]

Either the TM022 or TL022 should work

[PRR]

> "Mega". My drawing software does not take M instead..

The underlying SPICE functions take "m" or "M" for "milli". (It goes back to a day when only upper-case characters might be supported.)

The designation for "million" is "MEG", not "M'", not "Mega".

"M" works without complaint and gives you a number a billion times smaller than you expected. "Mega" should throw an error.

[Mr. Lime]

A friend of mine would like to have some sort of octave down option on this effect and so I would try to add an octave down part from the Robot Devil pedal.

I would be glad if someone could take a look at the octave down part and the mixing stage, if it's going to work?

This should be pretty similar to the Earthquake Devices Bit Commander if everything works well.
Is there a problem connecting the CD4024BE to +9V and the +9/-9V ground?

The CD4024 only takes 12V maximum..

[Mr. Lime]

Got anyone experience in mixing channels or octaves?

Anything I should change before I do a stripboard layout?

Of course I will post the layout here so that others could try the circuit too.